The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Heeger, P. S.
Right arrow Articles by Lehmann23, P. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Heeger, P. S.
Right arrow Articles by Lehmann23, P. V.
The Journal of Immunology, 2000, 165: 1278-1284.
Copyright © 2000 by The American Association of Immunologists

Comprehensive Assessment of Determinant Specificity, Frequency, and Cytokine Signature of the Primed CD8 Cell Repertoire Induced by a Minor Transplantation Antigen1

Peter S. Heeger2,*,{dagger}, Anna Valujskikh2,* and Paul V. Lehmann23{dagger}

* Department of Medicine, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, and {dagger} Institute of Pathology, Case Western Reserve University School of Medicine, Cleveland, OH 44106


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results and Discussion
 References
 
T cell immunity is often focused on one peptide segment of a complex protein Ag, with other epitopes inducing weaker, low frequency responses or no responses at all. Such determinant hierarchy has been well characterized for MHC class II-restricted CD4 cell immunity, but is less well understood for class I-restricted CD8 cell responses. We studied class I determinant recognition in a skin transplant model with ß-galactosidase (ß-gal) as a minor transplantation Ag. CD8 T cells from C57BL/6 mice that rejected congenic C57BL/6 ß-gal transgenic skin were tested in enzyme-linked immunospot assays for recall responses to single-step, overlapping, 9-mer peptides that spanned a 94-aa region of the ß-gal sequence. This approach provided every possible class I-restricted peptide for CD8 cell recognition, allowing us to define the in vivo frequency of CD8 cells specific for each of the 86 individual peptides. While four peptides were predicted to bind to the Kb or Db molecules, only one (ß-gal96–103) actually induced an immune response. No peptides outside of the motifs were recognized. Tolerization to ß-gal96–103 significantly prolonged ß-gal transgenic skin graft survival, confirming its immune dominance. Therefore, single-determinant dominance characterized this CD8 cell response. The data demonstrate the feasibility of large-scale, comprehensive, class I determinant mapping, an approach that should be indispensable in measuring CD8 cell immunity in humans.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results and Discussion
 References
 
During an infection, after vaccination, or during transplant rejection, naive Ag-specific T cells are induced to proliferate and expand clonally (1). The magnitude of the T cell response is defined by the numbers (frequencies) of memory T cells that are generated following the immune stimulus. Typically, several peptides on complex protein Ags have the potential to induce a T cell response (2, 3). Which of these peptides are actually immunogenic is determined by the MHC alleles expressed in an individual, by the number of precursor cells available for recognition of the respective peptide, and by the specific Ag-processing and presentation characteristics of the host (2, 3, 4, 5, 6). As each human codominantly expresses multiple HLA class I and class II loci, and each locus is highly polymorphic, a wide array of possible determinants can be recognized by that individual’s T cells. Moreover, as there is a vast diversity of HLA-allele combinations within the outbred human population, there is an essentially infinite array of peptides that can be recognized by T cells within the population as a whole.

Reliable algorithms are now available for predicting peptide binding to some (1, 2, 7, 8, 9, 10, 11, 12, 13, 14), but not most, murine or human MHC alleles. While such algorithms can predict peptide binding, they cannot predict which of these peptides are actually processed and presented, nor can they predict the relative strengths of the induced peptide-specific responses. Moreover, as T cell immunity to a given Ag is defined as the sum of the individual T cells present in the host responding to each of the antigenic peptides (restricted by the different MHC alleles), measurement of the response to a single peptide, i.e., by tetramer analysis, may not be an adequate reflection of the host’s immune status.

Numerous studies over the last 10 years have built our knowledge of MHC class II determinant hierarchy and recognition by CD4 cells in vivo (4, 6, 15, 16, 17). The progress was facilitated by the fact that exogenous Ag is readily processed for presentation on class II molecules (18, 19) and that the open ends of the peptide binding groove on class II molecules make them insensitive to frame shifts when long test peptides are used (20, 21). In addition, proliferation assays and, more recently, enzyme-linked immunospot (ELISPOT)4 assays, which require relatively few cells and little labor, have facilitated the assessment of CD4 immunity (6, 15, 22, 23). The published studies have clearly shown that frequently only one of several peptides with class II binding properties behaves in an immunodominant fashion, while other peptides are subdominant (they elicit weaker responses than the dominant determinant) or are cryptic (they do not elicit T cell responses) (6, 15, 16, 17). Moreover, these studies have shown that the determinant hierarchy can be dynamic, particularly under conditions of chronic immune pathology. T cell responses to determinants that initially behave as cryptic or subdominant can become increasingly prevalent and then critically contribute to the T cell-mediated inflammatory processes (16, 17). It is important to note that the dominance or crypticity of a given class II-restricted peptide cannot be predicted on the basis of the peptide’s ability to bind to MHC II molecules. In fact, the peptides exhibiting the strongest MHC binding ability can be the least immunogenic because they can negatively select high avidity T cell precursors with related specificity (24, 25, 26).

In contrast to the determinant hierarchy of CD4 cells, that of CD8 cells has proven to be a major challenge. Typically, CD8 T cell lines have been generated from immune hosts, a process that may favor the outgrowth of clones specific for some determinants over others and, hence, may not reflect the prevalence of CD8 cells specific for the different peptides in vivo (27). The readout system for CD8 immunity has primarily been the cytotoxicity (CTL) assay, which, in contrast to the proliferation assay, is not particularly sensitive, is rather labor intensive, and depends upon large numbers of effector cells (27, 28). Such limitations have impaired systematic studies aimed at establishing the frequency of CD8 memory cells in vivo capable of recognizing the different class I-restricted peptides of an Ag. It is only within the last several years that alternative methodologies have emerged, including MHC: peptide tetramer staining (29), intracytoplasmic cytokine staining (30, 31), and ELISPOT assays (32, 33, 34), that allow direct measurements of single Ag-specific CD8 cells ex vivo. The ELISPOT approach has been particularly useful for characterizing low frequency CD8 cell responses specific for known MHC class I-restricted peptides using freshly isolated material (35, 36, 37). Published studies have been confined to characterization a limited number of such peptides; however, systematic determinant mapping of CD8 responses has not been reported.

The MHC class I-processing machinery predominantly produces peptide epitopes from Ags synthesized within the APC (18, 19). Functional readout systems for the recognition of well-characterized model Ags by CD8 cells have, therefore, previously required that the Ag be introduced into the cytoplasm of the APC by transfection, viral infection, or osmotic shock (27, 35, 38). The standard approach for determinant mapping of class II determinants has involved screening with long peptides (up to 100 residues) followed by analysis of peptide series synthesized with 5- to 10-residue overlaps. These types of experiments have been feasible because the MHC II binding groove is open on both ends (39) so as to allow even long peptides to bind to MHC and to elicit detectable recall T cell immune responses. Analogous experiments aimed at identifying class I determinants have also been somewhat successful in defining determinant hierarchies in selected model systems (27, 40, 41, 42). However, the fact that the MHC class I peptide binding groove is closed on both ends and is thus relatively intolerant to frame shifts (43, 44) raises questions as to whether such screening approaches, using panels of peptides with 5- to 10-residue overlaps, might inadvertently miss relevant determinants. Theoretically, comprehensive class I determinant mapping requires the use of peptides of a length that can directly bind to the class I molecule, 8–10 aa (2, 3, 8), and that cover the molecule in steps of single amino acids (to bypass the intolerance to frame shifts). Even for short protein Ags, this approach would require up to hundreds of different peptides to be tested simultaneously with freshly isolated cells obtained from a single individual, a task that is not easily accomplished with CTL assays, tetramer technology, or intracytoplasmic cytokine staining.

In the present studies we used the ELISPOT approach to demonstrate the feasibility of such comprehensive measurements of class I determinant recognition in freshly isolated CD8 cells. This approach should promote the understanding of CD8 cell immunity in a variety of highly defined murine models and should facilitate the assessment of CD8 cell immunity in humans.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results and Discussion
 References
 
Animals

Male and female C57BL/6 (B6, H-2b), C57BL/6J-TgR(ROSA26)26 (ß-gal tg, H-2b), and C57BL/6-Ifngtm1Ts (B6 IFN-{gamma}-/-H-2b) animals, aged 6–8 wk, were purchased from The Jackson Laboratory (Bar Harbor, ME) and maintained in the specific pathogen-free animal facility at the Louis Stokes Cleveland Department of Veterans Affairs Medical Center. All the transgenic and knockout mice were backcrossed for >10 generations onto the B6 background.

Peptides

Peptides ß-gal96–103 (DAPIYTNV) and H-Ypb (WMHHNMDLI) were synthesized by Research Genetics (Huntsville, AL). ß-Gal106–114 (PITVNPPFV), ß-gal139–147 (TRIIFDGVN), and ß-gal178–186 (RAGENRLAN) were synthesized by Princeton Biomolecules (Columbus, OH). All these peptides were purified by HPLC to >98% purity. The single-step overlapping 9-mer peptides from the ß-gal region 93–187 were synthesized using the Multipin Peptide Synthesis System from Chiron Technologies (Raleigh, NC) and ranged in purity from 70 to 85%.

Immunizations

ß-Gal peptides were mixed with CFA (Life Technologies, Grand Island, NY) at a final concentration of 0.5 mg/ml, and 100 µl was injected s.c. into the flanks of the recipient mice. Intravenous tolerance was induced by single injections of 1 mg of peptide into the tail vein.

Placement and evaluation of skin grafts

Full-thickness trunk skin allografts were placed using standard techniques (45, 46). Skin was harvested from killed donor mice, cut into 0.5-cm2 pieces, and placed in sterile PBS until used for transplantation (<30 min). Recipient mice were anesthetized with pentobarbital (50 µg/g body weight) and shaved around the chest and abdomen. The skin graft was placed in a slightly larger graft bed prepared over the chest of the recipient and secured using Vaseline gauze and a bandage. Bandages were removed on day 7, and the grafts were then visually scored daily for evidence of rejection. The graft was considered fully rejected when it was >90% necrotic. In selected animals, allograft rejection was confirmed by histologic analysis.

T cell subset isolation

Splenic and lymph node CD8+ T cells were isolated using commercially available murine T cell isolation columns from R&D Systems (Minneapolis, MN) following the instructions supplied by the manufacturer. Resultant cells were washed in HBSS medium, counted by trypan blue exclusion, and resuspended at appropriate concentrations for use in the various assays. The purified T cell subpopulation was shown to be >92% CD8+ by flow cytometry.

ELISPOT assays

ELISPOT plates (ImmunoSpot P200, Cellular Technologies, Cleveland, OH) were coated with capture Abs diluted in PBS and then blocked for 1 h with sterile PBS/1% BSA and washed with sterile PBS (32, 33, 34, 45, 46). R46A2, produced and isolated in our laboratory from a hybridoma, was used at 4 µg/ml for IFN-{gamma}. Anti-IL-2 capture Ab (3 µg/ml; JES6-1A12, PharMingen, La Jolla, CA) was used for IL-2. 11B11 was produced and isolated in our laboratory from a hybridoma and was used at 2 µg/ml for IL-4. Anti-IL-5 capture Ab (5 µg/ml; TRFK4, PharMingen, La Jolla, CA) was used for IL-5. Various dilutions of purified CD8+ T cells (0.2–20 x 105/ml) in 200 µl of HL-1 medium (BioWhittaker, Walkersville, MD) were placed in each well with or without peptide plus spleen cell stimulators as APCs (in duplicate) and incubated at 37°C for 24 h in 5% CO2. After washing with PBS followed by PBS/0.025% Tween, detection Abs were added overnight. Biotin rat anti-mouse (XMG1.2, PharMingen, San Diego, CA; 4 µg/ml) was used for IFN-{gamma}, rat anti-mouse IL-2-biotin (2 µg/ml; JES6-5H4, PharMingen) was used for IL-2, rat anti-mouse IL-4-biotin (2 µg/ml; BVD4-1D11, PharMingen), and rat anti-mouse IL-5 (4 µg/ml TRFK5, PharMingen) was used for IL-5. Third reagents were added for 2 h at room temperature. Alkaline phosphatase-conjugated anti-biotin Ab (Vector, Burlingame CA) was used for IFN-{gamma} and IL-5, and streptavidin-HRP (Dako, Carpenteria, CA; 1/2000 in PBS/0.025% Tween) was used for IL-2 and IL-4. The IFN-{gamma} and IL-5 plates were developed using a nitro blue tetrazolium chloride (Bio-Rad Laboratories, Hercules, CA) and 5-bromo-4-chloro-3-indolyl phosphate (Sigma, St. Louis, MO) substrate. Sixty-six microliters of 60 mM nitro blue tetrazolium chloride in 70% dimethylformamide (DMF) plus 33 µl of 250 mM 5-bromo-4-chloro-3-indolyl phosphate in 100% DMF were dissolved in 10 ml of 0.1 M Trizma base, 0.1 M NaCl, and 0.1 M MgCl2 (pH 9.5); 200 µl of this mixture was placed in each ELISPOT well. The IL-2 and IL-4 spots were developed using 800 µl of 3-amino-9-ethylcarbazole (Pierce, Rockford, IL; 10 mg dissolved in 1 ml of DMF) mixed in 24 ml of 0.1 M sodium acetate, pH 5.0, plus 12 µl H202 (200 µl/well). The resulting spots were counted on a computer-assisted ELISPOT image analyzer (ImmunoSpot Series I Analyzer, Cellular Technologies, Cleveland, OH), which is designed to detect ELISPOTs based on size, shape, and colorimetric density (32, 33, 34, 45, 46).


   Results and Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
References
 
Experimental design

To study CD8 cell recognition of MHC class I-restricted determinants, we used a transplant model in which skin grafts from ß-gal tg mice on the C57BL/6 (B6) background (H-2b) were placed onto wild-type B6 recipients. In this model system, ß-gal serves as the sole minor transplantation Ag and causes rejection within 25 days after placement of the graft (46). The transgene is expressed under a ubiquitin promotor (47), leading to synthesis of a nonsecreted protein in the cytoplasm of donor skin cells and resulting in MHC class I-restricted and class II-restricted presentation of ß-gal-derived peptides (46). While both CD4 and CD8 cells seem to contribute to the rejection process, the CD8 compartment is essential, since B6 congenic CD8 knockout mice show greatly impaired and delayed rejection of B6 ß-gal tg skin (46). Thus, in the system used, a defined model Ag is not only expressed such that it can give rise to an Ag-specific class I-restricted CD8 cell response; in addition, the functional consequences of the induced CD8 response can be studied by rejection of the graft. The immune response was induced by the transplantation process itself, and purified CD8 cells were then tested for recognition of ß-gal-derived peptides when rejection was complete (days 25–30).

We synthesized a panel of 86 peptides spanning a segment of ß-gal, 93–187, previously implicated as containing immunodominant determinant(s) (46). The peptides were synthesized at 9 aa in length such that they could directly bind extracellularly to the MHC molecule when added as soluble peptides (without the requirement for further Ag processing). The 9-mer peptides spanned the protein sequence in steps of single amino acids (Fig. 1Go), and each peptide was then tested individually. This approach made the ex vivo readout system independent of Ag processing because every conceivable nonamer determinant was made available for MHC I binding and, hence, for recognition by the in vivo primed CD8 cells. If primed CD8 cells reactive to a given peptide were present in the host, they were stimulated in the recall assay.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 1. Schematic representation of single-step, overlapping, 9-mer peptides derived from the ß-gal sequence used for MHC class I determinant mapping. Standard single-letter codes for amino acids are used.

 
We used the cytokine ELISPOT approach to detect the peptide-reactive T cells, and to measure their frequency directly ex vivo (32, 33, 34, 45, 46). Naive T cells do not produce IFN-{gamma} or other effector cytokines upon initial Ag encounter, but acquire this ability through Ag-driven differentiation in vivo (34, 48). CD8 cells can differentiate into either TC1 cells, which are programmed to produce IFN-{gamma} and are cytolytic, or TC2 cells, which are programmed to secrete type 2 cytokines (IL-4, IL-5) and whose ability to kill is controversial (49, 50). We measured IFN-{gamma}, IL-2, IL-4, and IL-5 production to account for these different types of CD8 cell responses. Freshly isolated splenic CD8 cells were tested for peptide-induced cytokine production over 24 h, an interval too short for in vitro cytokine differentiation or cellular proliferation to occur (34, 51). The recall Ag-induced ELISPOTs reflect cognate cytokine production by individual Ag-stimulated T cells; therefore, the number of peptide-induced ELISPOTs reflected the number of the peptide-reactive T cells present in the test cell population that were preprogrammed in vivo to secrete the cytokine upon Ag challenge. The assay was performed with purified CD8 cells to ascertain the cellular source of cytokine production. For IFN-{gamma} measurements, the cytokine that dominated this TC1 recall response (46), spleen cell APCs were obtained from naive, congenic, IFN-{gamma} knockout mice to further assure that only cognate, CD8 cell-derived IFN-{gamma} was measured.

Single-determinant dominance of ß-gal recognition in CD8 cells

When we tested CD8 cells from mice that rejected ß-gal tg skin grafts for peptide-induced IFN-{gamma} production, we found that of the 86 simultaneously tested peptides, two adjacent peptides induced cytokine production (ß-gal95–103 and ß -gal96–104; Fig. 2Go) with a stimulation index of ~200 (212 spots/million for ß-gal95–103 vs less than two spots in medium alone or with the other ß-gal peptides). Naive B6 mice did not respond to the peptides or to ß-gal tg spleen cells (not shown). These data suggest that the peptides 95–103 and 96–104 have induced clonal sizes of IFN-{gamma}-producing memory cells at least 200 times larger than any of the other tested ß-gal peptides. The frequency of CD8 cells responding to intact ß-gal tg spleen cells (~150/million; Fig. 2Go) was similar to the frequency of CD8 cells responding to the peptides 95–103 and 96–104. This clearly identifies the 96–103 region as immune dominant, encompassing the overwhelming majority of the ß-gal-reactive CD8 cells. Consistent with our previously reported studies documenting that the recall response following ß-gal tg skin graft rejection is TC1 dominated (46), IL-5 ELISPOT production was not detectable over background (less than five spots per well) for the same peptide series (not shown). These results were invariably seen in five individual experiments using the entire peptide set. While we readily detected IFN-{gamma} producers specific for this region using our ELISPOT approach, we were not able to detect cytolytic activity toward any individual peptide determinant using standard ex vivo CTL experiments. We attribute this to the low frequency of peptide-reactive cells (Fig. 2Go), in the range of 200/million, which is below the known sensitivity of CTL assays (28). In confirmation of our findings, previously published studies by others, using an entirely different model system, demonstrated the immunogenicity of ß-gal96–103 (52). Our comprehensive analysis additionally shows, however, that ß-gal96–103 is not only immunogenic but also immune dominant; no other determinants within the tested region induced a detectable response.



View larger version (42K):
[in this window]
[in a new window]
  		FIGURE 2. Systematic mapping of ß-gal peptides recognized by CD8 cells using the 9-mer peptide series that spans the protein sequence in steps of single amino acids. Purified splenic CD8 T cells (>93% CD8+ by flow cytometry; not shown) were obtained from pooled spleens of three B6 mice following rejection of B6 ß-gal tg skin grafts (day 25 following placement of the graft) and were tested in individual IFN-{gamma} ELISPOT wells for responses to each of the 9-mer ß-gal peptides depicted in Fig. 1Go or to mitomycin C-treated ß-gal tg spleen cells. Two hundred thousand CD8 cells were placed in each well with 300,000 mitomycin C-treated spleen cell APCs from B6 IFN-{gamma}-/- mice plus 10 µM peptide. A, Representative individual IFN-{gamma} ELISPOT wells with the test peptide specified under each well. B, Frequency of detected spots for each peptide as determined by computer-assisted image analysis. The experiments were repeated four times with identical results. No IL-5 ELISPOTs were detectable for any peptide in the series (not shown).

 
To confirm independently our results and to characterize more fully the cytokine signature of the recall response, we separately synthesized an HPLC-purified form of ß-gal96–103 (the residues common to 95–103 and 96–104). Recall ELISPOT analysis using this peptide confirmed the specificity of the induced CD8 cells for ß-gal96–103 (Fig. 3Go). The TC1 cytokine signature of the response, characterized by IFN-{gamma} production without IL-2, IL-4, or IL-5 production, was noted over a 6-log-fold concentration of the peptide in the recall assay (0.001–100 µg/ml; not shown). No cytokine production in response to this peptide was detectable in unprimed mice or in mice that rejected fully MHC disparate skin grafts (not shown).



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 3. CD8 immunity following rejection of ß-gal transgenic skin grafts is IFN-{gamma} producing and is focused to one of four candidate determinants that fit the Kb/Db binding motifs. Purified splenic CD8 T cells (>93% CD8+ by flow cytometry; not shown) were obtained from pooled spleens of three B6 mice following rejection of B6 ß-gal tg skin grafts (day 25) and tested in cytokine ELISPOT wells for responses to HPLC-purified peptides ß-gal96–103, ß-gal106–114, ß-gal139–147, and ß-gal178–186. Frequencies of IFN-{gamma}-producing CD8 cells in response to the four candidate determinants are shown. Spot frequencies represent the mean values of duplicate wells (<10% variability between wells) and were titrated over a range of 0.001–100 µg/ml with similar results (not shown). No IL-2, IL-4, or IL-5 was detected in response to any of the peptides over this concentration range (not shown). The experiment was repeated with identical results.

 
Peptides fitting binding motifs are not all immunogenic

The binding motifs for both the Kb and the Db MHC molecules are shown in Table IGo. The peptide ß-gal96–103 fits the Kb motif in that it has the required anchor residues at the 5 and 8 positions, but is not fully compatible with the motif in that the peptide lacks the tyrosine residue in position 3 (Table IGo). We searched the amino acid sequence of ß-gal93–187 for other peptides that fit either of the Kb or the Db motif, both manually and with two epitope binding motif search programs available on the internet (14). In addition to ß-gal96–103, three other candidate peptides were identified (Table IGo). One of these, ß-gal139–147, manually fit the Kb motif equally well as 96–103 did, using acceptable anchor residues at positions 5 and 8. These two determinants (96–103 and 139–147) had the highest scores for binding to Kb based on one of the epitope search programs (14). Two other peptides (106–114 and 178–186) generally fit the Db motif, having appropriate anchor residues at positions 4 and 5. While these latter two peptides did not fully fit the Db motif, in that they lacked the usual binding residues at several other positions (Table IGo), they both came up as strong potential Db binders using one epitope search program (14). In fact, peptide 178–186 had the highest binding score of all peptides in the region studied. Interestingly, the second search program (53) yielded somewhat different results; while peptides 96–103 and 178–186 were predicted to be good binders to Kb and Db, respectively, 139–147 and 106–114 were ranked much lower. As shown above (Fig. 2Go), CD8 cell memory to any of these three alternative peptides (ß-gal106–114, ß-gal139–147, ß-gal178–186) was not detected with the peptide scan screening approach. With different well-established algorithms giving different results, and in face of recent studies that suggest that binding affinity does not suffice to predict actual immune dominance (54), the data underline the need for performing comprehensive mapping analyses.


View this table:
[in this window]
[in a new window]
  		Table I. ß-gal- and H-Y-derived determinants fitting the peptide binding motifs for MHC class I Kb and Db moleculesa

 
To test more stringently for the presence of T cells recognizing these three peptides, we repeated the transplant experiments using a different batch of newly synthesized and HPLC-purified peptides in the recall analysis. While ß-gal96–103 again induced a potent recall response, none of the other candidate determinant peptides induced an IFN-{gamma}, IL-2, IL-4, or IL-5 response over background (Fig. 3Go and data not shown), even when tested over a wide range of concentrations (0.001–100 µg/ml). We conclude that the binding motif predictions were useful in identifying the one determinant that was immune dominant following ß-gal tg skin graft rejection, yet this determinant was only one of four equally attractive candidate peptides. While we cannot formally state in our model whether the inability to detect recall immunity to determinants other than ß-gal96–103 following rejection was due to a lack of peptide binding to MHC or was due to other factors (for example, specific deficiencies in the T cell repertoire or in Ag processing), our data clearly show that the only determinant within the region tested that elicits a recall immune response following rejection is ß-gal96–103.

The rules of determinant hierarchy may vary with conditions of induction of the immune response (15, 16). We therefore tested whether these four peptides, all generally fitting the canonical motifs, would elicit immune responses when injected individually with CFA. Only ß-gal96–103 was immunogenic, inducing a vigorous IFN-{gamma}+ (IL-2-, IL-4-, IL-5-) recall response (Fig. 4Go). The CD8 memory cells induced by immunization with the peptide could also be activated by spleen cells obtained from ß-gal tg mice. Moreover, the frequencies of the IFN-{gamma}-producing CD8 T cells were essentially identical whether the recall response was elicited by the peptide or by the ß-gal tg cells (Fig. 4Go). The findings demonstrate that ß-gal96–103 is naturally processed and presented, thus formally defining it as an immune dominant determinant (6, 15, 16).



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 4. Immunization with ß-gal96–103, but not ß-gal106–114, ß-gal139–147, or ß-gal178–186, primes an IFN-{gamma}-producing immune response. Purified CD8 T cells obtained from draining LNs on day 10 following immunization with peptide in CFA were tested in recall IFN-{gamma} ELISPOT assays in response to peptide (at the concentrations depicted) or to mitomycin C-treated, ß-gal tg spleen cells (300,000/well). Responses to control peptides were not detectable over background (less than five spots per well, not shown). No IL-2, IL-4, or IL-5 was detected in response to any of the peptides over this concentration range (not shown). Spot frequencies represent the mean values of duplicate wells (<10% variability between wells). The findings are representative of two independent experiments.

 
Cryptic determinants are defined as those capable of inducing a peptide-specific response following immunization, but which cannot be elicited in a recall assay by the naturally processed protein itself (6, 15, 16). Immunization with ß-gal106–114, ß-gal139–147, or ß-gal178–186 did not elicit a cytokine recall response to any of the immunizing peptides themselves or to the ß-gal tg APCs (Fig. 4Go); thus, these three candidate peptides do not even qualify as cryptic determinants. Despite the binding motif predictions, either these peptides do not bind to Kb/Db or there are holes in the T cell repertoire that would otherwise recognize them. Regardless of the reason, the data clearly show that even when peptide binding motifs are known, predictions of determinant utilization cannot substitute for actual measurements of the induced T cell immune response.

Induction of tolerance to 96–103 prolongs ß-gal tg skin graft survival

To verify independently the central role of 96–103 in CD8 cell recognition of ß-gal, we induced tolerance by i.v. injection of a high dose (1 mg) of soluble peptide before placement of skin grafts. As shown in Fig. 5Go, injection of ß-gal96–103 more than doubled the time to rejection compared with that in untreated control animals, a statistically (p < 0.05) and biologically highly significant result. In contrast, preinjection of peptide ß-gal106–114 did not affect the rejection (Fig. 5GoA). To control for peptide specificity, we preinjected recipient mice with a peptide that fits the Db binding motif (H-Ypb, Table IGo) and whose immune dominance is well established (55). Pretreatment with this peptide (H-Ypb) did not affect the kinetics of ß-gal tg skin graft rejection (Fig. 5GoA), but did prolong survival of B6 male skin on Db-expressing female recipients (33.5 days for H-Ypb-injected vs 17.5 days for control mice; n = 4/group; data not shown). As a further specificity control, i.v. injection of ß-gal96–103 had no effect on the survival of third-party minor Ag-disparate skin grafts: H-Y-expressing male B6 skin grafts on female B6 recipients (Fig. 5GoB). Syngeneic female B6 skin grafts were not rejected (graft survival, >40 days; n = 3; not shown)



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 5. Induction of tolerance to ß-gal96–103 prolongs survival of ß-gal tg skin grafts. Full-thickness trunk skin grafts were placed on groups of recipient mice 12 days after i.v. injection of peptide. A, Survival of female B6 ß-gal tg skin grafts on female wild-type B6 recipients. B, Survival of male B6 skin grafts on female B6 recipients.

 
While i.v. injection of ß-gal96–103 specifically prolonged ß-gal tg skin graft survival, the grafts were ultimately rejected. T cell responses studied in these mice at the time of rejection (days 55–60) showed that ß-gal96–103-specific T cells producing IFN-{gamma} IL-2, IL-4, or IL-5 were not detectable (not shown). This confirmed that the i.v. injection tolerized the CD8 T cells specific for ß-gal96–103 and further showed that these T cells did not recover from the tolerant state and were not replenished from the thymus to mediate graft rejection. Moreover, systematic determinant mapping studies of the 93–187 region using CD8 and CD4 T cells did not reveal emergence of responses to any of the peptides over background. Within the region studied, therefore, there was no shifting of determinant recognition to compensate for the loss of the immune dominant response. In particular, when we looked for responses to the three additional candidate peptides, again titrating those peptides across a 6-log concentration, we did not detect any cytokine production (not shown). The data clearly show that ß-gal96–103 is a key determinant in the early CD8 cell-mediated rejection of ß-gal tg grafts.

To account for the late rejection, one would have to postulate that other CD8 responses outside of the region that we studied progressively gained prevalence so as to result ultimately in a rejection phenotype (i.e., determinant spreading), a finding that has frequently been observed for CD4 cell responses in other model systems (6, 15, 16, 17, 56). Alternatively, it is conceivable that the delayed rejection was driven by ß-gal-reactive CD4 cells. The emergence of CD4-mediated rejection in animals tolerized to ß-gal96–103 is supported by experiments in which congenic B6 CD8-/- mice (more than eight backcrosses to B6) rejected B6 ß-gal tg skin (38.8 days; n = 4) with a >2-wk delay compared with wild-type B6 recipients (21 days; n = 4). Moreover, we have previously published studies demonstrating ß-gal-specific delayed-type hypersensitivity, a predominantly CD4-mediated effector function, in animals undergoing rejection (46). Regardless of the final effector mechanisms that resulted in graft rejection, induction of tolerance to ß-gal96–103 changed the peptide determinant hierarchy. More extensive utilization of this systematic determinant mapping approach should permit further progress in understanding the dynamics of determinant utilization following tolerization.

Concluding remarks

Using ß-gal as a model Ag and skin transplantation as a method for priming CD8 cell immunity, we have demonstrated the feasibility of large-scale, class I determinant mapping directly ex vivo. While a motif search predicted four candidate peptides, we found that only one of these peptides, ß-gal96–103, actually functioned as a target of the CD8 cell attack. The immune dominance of this determinant was defined using functional recall assays and was verified by in vivo tolerization experiments resulting in prolonged allograft survival. While none of the peptides exactly fit the known MHC I binding motifs, our data confirm that such motifs (when known) can be useful starting points for narrowing down candidate peptides. More importantly, however, the technical advancement reported here should allow the complete assessment of CD8 cell repertoire specific for any Ag and restricted by any MHC class I allele, including those murine and human MHC I alleles for which binding motifs are not well characterized. Particularly for humans, in whom the essentially infinite numbers of class I allelic combinations have been an obstacle to the rational assessment of CD8 cell function, such systematic mapping should facilitate the understanding of CD8 cell immunity during infection, autoimmune disease, cancer, and transplant rejection and may aid in designing specific immunotherapies and vaccines.


   Acknowledgments
 
We thank Earla Biekert for her technical support and editorial assistance.


   Footnotes
 
1 This work was supported by research grants from the National Institutes of Health (DK48799, AI42635, and AI/DK44484 to P.V.L. and AI43578-01 to P.S.H.). P.S.H. is a recipient of a Clinical Scientist Award from the National Kidney Foundation. Back

2 P.S.H., A.V., and P.V.L. contributed equally to this work. Back

3 Address correspondence and reprint requests to Dr. Paul Lehmann, Institute of Pathology, 929 Biomedical Research Building, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106-4943. Back

4 Abbreviations used in this paper: ELISPOT, enzyme-linked immunospot; ß-gal, ß-galactosidase; tg, transgenic; DMF, dimethylformamide. Back

Received for publication December 3, 1999. Accepted for publication May 16, 2000.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
 References
 

  1. Parker, K., and R. Martin. 1996. Peptide binding motifs for MHC class I and II molecules. In Current Protocols in Immunology, Vol. 3. J. Coligan, A. Kruisbeek, D. Margulies, E. Shevach, and W. Strober, eds. John Wiley & Sons, New York, p. A.1I1.
  2. Falk, K., O. Rotzschke, S. Stevanovic, G. Jung, H. G. Rammensee. 1991. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 351:290.[Medline]
  3. Rotzschke, O., K. Falk, K. Deres, H. Schild, M. Norda, J. Metzger, G. Jung, H. G. Rammensee. 1990. Isolation and analysis of naturally processed viral peptides as recognized by cytotoxic T cells. Nature 348:252.[Medline]
  4. Gammon, G., N. Shastri, J. Cogswell, S. Wilbur, S. Sadegh-Nasseri, U. Krzych, A. Miller, E. Sercarz. 1987. The choice of T-cell epitopes utilized on a protein antigen depends on multiple factors distant from, as well as at the determinant site. Immunol. Rev. 98:53.[Medline]
  5. Gammon, G., H. M. Geysen, R. J. Apple, E. Pickett, M. Palmer, A. Ametani, E. E. Sercarz. 1991. T cell determinant structure: cores and determinant envelopes in three mouse major histocompatibility complex haplotypes. J. Exp. Med. 173:609.[Abstract/Free Full Text]
  6. Sercarz, E. E., P. V. Lehmann, A. Ametani, G. Benichou, A. Miller, K. Moudgil. 1993. Dominance and crypticity of T cell antigenic determinants. Annu. Rev. Immunol. 11:729.[Medline]
  7. Engelhard, V. H.. 1994. Structure of peptides associated with class I and class II MHC molecules. Annu. Rev. Immunol. 12:181.[Medline]
  8. Maier, R., K. Falk, O. Rotzschke, B. Maier, V. Gnau, S. Stevanovic, G. Jung, H. G. Rammensee, A. Meyerhans. 1994. Peptide motifs of HLA-A3, -A24, and -B7 molecules as determined by pool sequencing. Immunogenetics 40:306.[Medline]
  9. Parker, K. C., M. A. Bednarek, L. K. Hull, U. Utz, B. Cunningham, H. J. Zweerink, W. E. Biddison, J. E. Coligan. 1992. Sequence motifs important for peptide binding to the human MHC class I molecule, HLA-A2. J. Immunol. 149:3580.[Abstract]
  10. Ruppert, J., J. Sidney, E. Celis, R. T. Kubo, H. M. Grey, A. Sette. 1993. Prominent role of secondary anchor residues in peptide binding to HLA-A2.1 molecules. Cell 74:929.[Medline]
  11. Sette, A., S. Buus, S. Colon, C. Miles, H. M. Grey. 1988. I-Ad-binding peptides derived from unrelated protein antigens share a common structural motif. J. Immunol. 141:45.[Abstract]
  12. Sette, A., S. Buus, E. Appella, J. A. Smith, R. Chesnut, C. Miles, S. M. Colon, H. M. Grey. 1989. Prediction of major histocompatibility complex binding regions of protein antigens by sequence pattern analysis. Proc. Natl. Acad. Sci. USA 86:3296.[Abstract/Free Full Text]
  13. Sigal, L. J., P. Goebel, D. E. Wylie. 1995. Db-binding peptides from influenza virus: effect of non-anchor residues on stability and immunodominance. Mol. Immunol. 32:623.[Medline]
  14. Rammensee, H. 2000. SYFPEITHI epitope prediction website. http://134.2.96.221/scripts/hlaserver.dll/EpPredict.htm.
  15. Lehmann, P. V., T. Forsthuber, A. Miller, E. E. Sercarz. 1992. Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen. Nature 358:155.[Medline]
  16. Lehmann, P. V., E. E. Sercarz, T. Forsthuber, C. M. Dayan, G. Gammon. 1993. Determinant spreading and the dynamics of the autoimmune T-cell repertoire. Immunol. Today 14:203.[Medline]
  17. McRae, B. L., C. L. Vanderlugt, M. C. Dal Canto, S. D. Miller. 1995. Functional evidence for epitope spreading in the relapsing pathology of experimental autoimmune encephalomyelitis. J. Exp. Med. 182:75.[Abstract/Free Full Text]
  18. Unanue, E. R., J. C. Cerottini. 1989. Antigen presentation. FASEB J. 3:2496.[Abstract]
  19. Harding, C. V., E. R. Unanue. 1990. Cellular mechanisms of antigen processing and the function of class I and II major histocompatibility complex molecules. Cell Regul. 1:499.[Medline]
  20. Dessen, A., C. M. Lawrence, S. Cupo, D. M. Zaller, D. C. Wiley. 1997. X-ray crystal structure of HLA-DR4 (DRA*0101, DRB1*0401) complexed with a peptide from human collagen II. Immunity 7:473.[Medline]
  21. Fremont, D. H., D. Monnaie, C. A. Nelson, W. A. Hendrickson, E. R. Unanue. 1998. Crystal structure of I-Ak in complex with a dominant epitope of lysozyme. Immunity 8:305.[Medline]
  22. Jager, E., D. Jager, J. Karbach, Y. T. Chen, G. Ritter, Y. Nagata, S. Gnjatic, E. Stockert, M. Arand, L. J. Old, et al 2000. Identification of NY-ESO-1 epitopes presented by human histocompatibility antigen (HLA)-DRB4*0101-0103 and recognized by CD4+ T lymphocytes of patients with NY-ESO-1-expressing melanoma. J. Exp. Med. 191:625.[Abstract/Free Full Text]
  23. Anderson, R. P., P. Degano, A. J. Godkin, D. P. Jewell, A. V. Hill. 2000. In vivo antigen challenge in celiac disease identifies a single transglutaminase-modified peptide as the dominant A-gliadin T-cell epitope. Nat. Med. 6:337.[Medline]
  24. Ashton-Rickardt, P. G., L. Van Kaer, T. N. Schumacher, H. L. Ploegh, S. Tonegawa. 1993. Peptide contributes to the specificity of positive selection of CD8+ T cells in the thymus. Cell 73:1041.[Medline]
  25. Ashton-Rickardt, P. G., S. Tonegawa. 1994. A differential-avidity model for T-cell selection. Immunol. Today 15:362.[Medline]
  26. Targoni, O. S., P. V. Lehmann. 1998. Endogenous myelin basic protein inactivates the high avidity T cell repertoire. J. Exp. Med. 187:2055.[Abstract/Free Full Text]
  27. Yewdell, J. W., J. R. Bennink. 1999. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immunol. 17:51.[Medline]
  28. McMichael, A. J., C. A. O’Callaghan. 1998. A new look at T cells. J. Exp. Med. 187:1367.[Free Full Text]
  29. McHeyzer-Williams, M. G., J. D. Altman, M. M. Davis. 1996. Enumeration and characterization of memory cells in the TH compartment. Immunol. Rev. 150:5.[Medline]
  30. Openshaw, P., E. E. Murphy, N. A. Hosken, V. Maino, K. Davis, K. Murphy, A. O’Garra. 1995. Heterogeneity of intracellular cytokine synthesis at the single-cell level in polarized T helper 1 and T helper 2 populations. J. Exp. Med. 182:1357.[Abstract/Free Full Text]
  31. Elson, L. H., T. B. Nutman, D. D. Metcalfe, C. Prussin. 1995. Flow cytometric analysis for cytokine production identifies T helper 1, T helper 2, and T helper 0 cells within the human CD4+CD27- lymphocyte subpopulation. J. Immunol. 154:4294.[Abstract]
  32. Forsthuber, T., H. C. Yip, P. V. Lehmann. 1996. Induction of TH1 and TH2 immunity in neonatal mice. Science 271:1728.[Abstract]
  33. Matesic, D., A. Valujskikh, E. Pearlman, A. W. Higgins, A. C. Gilliam, P. S. Heeger. 1998. Type 2 immune deviation has differential effects on alloreactive CD4+ and CD8+ T cells. J. Immunol. 161:5236.[Abstract/Free Full Text]
  34. Matesic, D., P. V. Lehmann, P. S. Heeger. 1998. High-resolution characterization of cytokine-producing alloreactivity in naive and allograft-primed mice. Transplantation 65:906.[Medline]
  35. Larsson, M., X. Jin, B. Ramratnam, G. S. Ogg, J. Engelmayer, M. A. Demoitie, A. J. McMichael, W. I. Cox, R. M. Steinman, D. Nixon, et al 1999. A recombinant vaccinia virus based ELISPOT assay detects high frequencies of Pol-specific CD8 T cells in HIV-1-positive individuals. AIDS 13:767.[Medline]
  36. Kaul, R., F. A. Plummer, J. Kimani, T. Dong, P. Kiama, T. Rostron, E. Njagi, K. S. MacDonald, J. J. Bwayo, A. J. McMichael, et al 2000. HIV-1-specific mucosal CD8+ lymphocyte responses in the cervix of HIV-1-resistant prostitutes in Nairobi. J. Immunol. 164:1602.[Abstract/Free Full Text]
  37. Yang, J., V. M. Lemas, I. W. Flinn, C. Krone, R. F. Ambinder. 2000. Application of the ELISPOT assay to the characterization of CD8+ responses to Epstein-Barr virus antigens. Blood 95:241.[Abstract/Free Full Text]
  38. Moore, M. W., F. R. Carbone, M. J. Bevan. 1988. Introduction of soluble protein into the class I pathway of antigen processing and presentation. Cell 54:777.[Medline]
  39. Fremont, D. H., W. A. Hendrickson, P. Marrack, J. Kappler. 1996. Structures of an MHC class II molecule with covalently bound single peptides. Science 272:1001.[Abstract]
  40. Horwitz, M. S., Y. Yanagi, M. B. Oldstone. 1994. T-cell receptors from virus-specific cytotoxic T lymphocytes recognizing a single immunodominant nine-amino-acid viral epitope show marked diversity. J. Virol. 68:352.[Abstract/Free Full Text]
  41. Evans, D. T., M. S. Piekarczyk, T. M. Allen, J. E. Boyson, M. Yeager, A. L. Hughes, F. M. Gotch, V. S. Hinshaw, D. I. Watkins. 1997. Immunodominance of a single CTL epitope in a primate species with limited MHC class I polymorphism. J. Immunol. 159:1374.[Abstract]
  42. Wallace, M. E., R. Keating, W. R. Heath, F. R. Carbone. 1999. The cytotoxic T-cell response to herpes simplex virus type 1 infection of C57BL/6 mice is almost entirely directed against a single immunodominant determinant. J. Virol. 73:7619.[Abstract/Free Full Text]
  43. Bjorkman, P. J., M. A. Saper, B. Samraoui, W. S. Bennett, J. L. Strominger, D. C. Wiley. 1987. Structure of the human class I histocompatibility antigen, HLA-A2. Nature 329:506.[Medline]
  44. Bjorkman, P. J., P. Parham. 1990. Structure, function, and diversity of class I major histocompatibility complex molecules. Annu. Rev. Biochem. 59:253.[Medline]
  45. Valujskikh, A., D. Matesic, A. Gilliam, D. Anthony, T. M. Haqqi, P. S. Heeger. 1998. T cells reactive to a single immunodominant self-restricted allopeptide induce skin graft rejection in mice. J. Clin. Invest. 101:1398.[Medline]
  46. Valujskikh, A., D. Matesic, P. S. Heeger. 1999. Characterization and manipulation of T cell immunity to skin grafts expressing a transgenic minor antigen. Transplantation 68:1029.[Medline]
  47. Friedrich, G., P. Soriano. 1991. Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice. Genes Dev. 5:1513.[Abstract/Free Full Text]
  48. Ehlers, S., K. A. Smith. 1991. Differentiation of T cell lymphokine gene expression: the in vitro acquisition of T cell memory. J. Exp. Med. 173:25.[Abstract/Free Full Text]
  49. Sad, S., R. Marcotte, T. R. Mosmann. 1995. Cytokine-induced differentiation of precursor mouse CD8+ T cells into cytotoxic CD8+ T cells secreting Th1 or Th2 cytokines. Immunity 2:271.[Medline]
  50. Mosmann, T. R., L. Li, S. Sad. 1997. Functions of CD8 T-cell subsets secreting different cytokine patterns. Semin. Immunol. 9:87.[Medline]
  51. Heeger, P. S., N. S. Greenspan, S. Kuhlenschmidt, C. Dejelo, D. E. Hricik, J. A. Schulak, M. Tary-Lehmann. 1999. Pretransplant frequency of donor-specific, IFN-{gamma}-producing lymphocytes is a manifestation of immunologic memory and correlates with the risk of posttransplant rejection episodes. J. Immunol. 163:2267.[Abstract/Free Full Text]
  52. Overwijk, W. W., D. R. Surman, K. Tsung, N. P. Restifo. 1997. Identification of a Kb-restricted CTL epitope of ß-galactosidase: potential use in development of immunization protocols for "self" antigens. Methods 12:117.[Medline]
  53. Parker, K. 2000. HLA Peptide Binding Predictions. National Institutes of Health. http://bimas.dcrt.nih.gov:80/molbio/hla_bind/.
  54. Chen, W., L. C. Anton, J. R. Bennink, J. W. Yewdell. 2000. Dissecting the multifactorial causes of immunodominance in class I-restricted T cell responses to viruses. Immunity 12:83.[Medline]
  55. Simpson, E., D. Scott, P. Chandler. 1997. The male-specific histocompatibility antigen, H-Y: a history of transplantation, immune response genes, sex determination and expression cloning. Annu. Rev. Immunol. 15:39.[Medline]
  56. Kaufman, D. L., M. Clare-Salzler, J. Tian, T. Forsthuber, G. S. Ting, P. Robinson, M. A. Atkinson, E. E. Sercarz, A. J. Tobin, P. V. Lehmann. 1993. Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 366:69.[Medline]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
P. J. Wysocki, E. P. Kwiatkowska, U. Kazimierczak, W. Suchorska, D. W. Kowalczyk, and A. Mackiewicz
Captopril, an Angiotensin-Converting Enzyme Inhibitor, Promotes Growth of Immunogenic Tumors in Mice.
Clin. Cancer Res., July 1, 2006; 12(13): 4095 - 4102.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
A. Valujskikh, Q. Zhang, and P. S. Heeger
CD8 T Cells Specific for a Donor-Derived, Self-Restricted Transplant Antigen Are Nonpathogenic Bystanders after Vascularized Heart Transplantation in Mice
J. Immunol., February 15, 2006; 176(4): 2190 - 2196.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
C. He, S. Schenk, Q. Zhang, A. Valujskikh, J. Bayer, R. L. Fairchild, and P. S. Heeger
Effects of T Cell Frequency and Graft Size on Transplant Outcome in Mice
J. Immunol., January 1, 2004; 172(1): 240 - 247.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
B. Pantenburg, F. Heinzel, L. Das, P. S. Heeger, and A. Valujskikh
T Cells Primed by Leishmania major Infection Cross-React with Alloantigens and Alter the Course of Allograft Rejection
J. Immunol., October 1, 2002; 169(7): 3686 - 3693.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Heeger, P. S.
Right arrow Articles by Lehmann23, P. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Heeger, P. S.
Right arrow Articles by Lehmann23, P. V.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS